The Constitutional Controller: Doubt-Calibrated Steering of Compliant Agents

1Artificial Intelligence and Machine Learning Lab - TU Darmstadt
2Honda Research Institute Europe
3Uncertainty in Artificial Intelligence Group - TU Eindhoven
4Hessian AI    5Centre for Cognitive Science    6DFKI
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2026

CoCo integrates neuro-symbolic reasoning and doubt modeling to steer autonomous agents towards safety and compliance.

Abstract

Ensuring reliable and rule-compliant behavior of autonomous agents in uncertain environments remains a fundamental challenge in modern robotics. Our work shows how neuro-symbolic systems, which integrate probabilistic, symbolic white-box reasoning models with deep learning methods, offer a powerful solution to this challenge. They enable the simultaneous consideration of explicit rules and neural models trained on noisy data, combining the strengths of structured reasoning with flexible representations. To this end, we introduce the Constitutional Controller (CoCo), a novel framework designed to enhance the safety and reliability of agents by reasoning over deep probabilistic logic programs that represent constraints, such as those found in shared traffic spaces. Furthermore, we propose the concept of self-doubt, implemented as a probability density conditioned on doubt features such as travel velocity, employed sensors, or health factors. In a real-world aerial mobility study, we demonstrate CoCo's advantages for intelligent autonomous systems to learn appropriate doubts and navigate complex, uncertain environments safely and compliantly.

BibTeX

@inproceedings{kohaut2026coco,
  title={The Constitutional Controller: Doubt-Calibrated Steering of Compliant Agents},
  author={Simon Kohaut and Felix Divo and Navid Hamid and Benedict Flade and Julian Eggert and Devendra Singh Dhami and Kristian Kersting},
  booktitle={IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  year={2026}
}